World Congress 2026 Europe - Virtual Stage • Jul 2, 2026 • Session details

Your AI Agent is just a while loop with an API call. Let me prove it

Michał Michalczuk

Your complex AI agent is just a stateless language model trapped inside a while loop. Strip away the abstractions to master agent architecture, token economics, and sandbox security.

Pause
Mute Enter Fullscreen
#1 about 3 min

Exploring modern AI SDKs and production deployment challenges

Modern AI SDKs simplify initial creation but often introduce unexpected costs and compliance hurdles during production.

#2 about 2 min

Building a basic conversational AI agent in Node.js

A rudimentary script passes persistent conversation history arrays to an API to simulate chat memory.

#3 about 3 min

Integrating tool definitions for local bash shell execution

Providing a shell capability lets the agent directly execute terminal process commands on the host machine.

#4 about 3 min

Securing tool execution using isolated Docker sandboxes

Running unpredictable shell commands securely requires isolated sandbox environments to protect base operating systems.

#5 about 4 min

Implementing persistent memory through self-modifying agent capabilities

Enabling self-modification allows an agent to persist important conversational logic and references in local storage files.

#6 about 3 min

Building a multi-agent system with coding sub-agents

Separating complex tasks into specialized sub-agents minimizes context payloads and safely orchestrates delegated coding workflows.

#7 about 3 min

Defining the core architecture of an AI agent

The underlying pattern of commercial AI products relies on a continuous loop connecting foundation models with robust tool harnesses.

#8 about 7 min

Analyzing token costs and cumulative history payloads

Stateless requests force consecutive accumulation of context history and static schema structures, massively inflating input token consumption.

#9 about 5 min

Measuring real-world token consumption in data-intensive workflows

Retrieving background context from broad domains like calendars rapidly exhausts token limits while expanding request dimensions.

#10 about 3 min

Key architectural takeaways for building reliable AI agents

Designing helpful agent experiences hinges on deeply understanding stateless API behaviors and strictly defining task boundaries.

Matching moments

2:36 min

Exploring the core architecture and components of AI agents

Jörg Neumann Jörg Neumann · World Congress 2025

4:20 min

Understanding the core AI agent execution loop

Brad Axen Brad Axen · World Congress 2025

3:25 min

Constructing scalable AI solutions using LangChain and LangGraph

Julián Duque Julián Duque · World Congress 2025

4:02 min

Core architecture and reasoning loops of AI agents

Benjamin Smith Benjamin Smith · World Congress 2026 Europe

2:04 min

Inspiration and challenges of scaling AI agent communication

diabhey diabhey · World Congress 2025

1:22 min

Leveraging the comprehensive generative artificial intelligence stack

Duan Lightfoot Duan Lightfoot · World Congress 2024